Conceptual-level workflow modeling of scientific experiments using NMR as a case study

Foundational Practices of Research Data Management

The importance of research data has grown as researchers across disciplines seek to ensure reproducibility, facilitate data reuse, and acknowledge data as a valuable scholarly commodity. Researchers are under increasing pressure to share their data for validation and reuse. Adopting good data management practices allows researchers to efficiently locate their data, understand it, and use it throughout all of the stages of a project and in the future. Additionally, good data management can streamline data analysis, visualization, and reporting, thus making publication less stressful and time-consuming. By implementing foundational practices of data management, researchers set themselves up for success by formalizing processes and reducing common errors in data handling, which can free up more time for research. This paper provides an introduction to best practices for managing all types of data.

Semantic Mediation to Improve Reproducibility for Biomolecular NMR Analysis

Two barriers to computational reproducibility are the ability to record the critical metadata required for rerunning a computation, as well as translating the semantics of the metadata so that alternate approaches can easily be configured for verifying computational reproducibility. We are addressing this problem in the context of biomolecular NMR computational analysis by developing a series of linked ontologies which define the semantics of the various software tools used by researchers for data transformation and analysis. Building from a core ontology representing the primary observational data of NMR, the linked data approach allows for the translation of metadata in order to configure alternate software approaches for given computational tasks. In this paper we illustrate the utility of this with a small sample of the core ontology as well as tool-specific semantics for two third-party software tools. This approach to semantic mediation will help support an automated approach to validating the reliability of computation in which the same processing workflow is implemented with different software tools. In addition, the detailed semantics of both the data and the processing functionalities will provide a method for software tool classification.

Curating Scientific Workflows for Biomolecular Nuclear Magnetic Resonance Spectroscopy

This paper describes our recent and ongoing efforts for enhancing the curation of scientific workflows to improve reproducibility and reusability of biomolecular nuclear magnetic resonance (bioNMR) data. Our efforts have focused on both developing a workflow management system, called CONNJUR Workflow Builder (CWB), as well as refactoring our workflow data model to make use of the PREMIS model for digital preservation. This revised workflow management system will be available through the NMRbox cloud-computing platform for bioNMR. In addition, we are implementing a new file structure which bundles the original binary data files along with PREMIS XML records describing the provenance of the data. These are packaged together using a standardized file archive utility. In this manner, the provenance and data curation information is maintained together along with the scientific data. The benefits and limitations of these approaches as well as future directions are discussed.

CONNJUR Workflow Builder: a software integration environment for spectral reconstruction

CONNJUR Workflow Builder (WB) is an open-source software integration environment that leverages existing spectral reconstruction tools to create a synergistic, coherent platform for converting biomolecular NMR data from the time domain to the frequency domain. WB provides data integration of primary data and metadata using a relational database, and includes a library of pre-built workflows for processing time domain data. WB simplifies maximum entropy reconstruction, facilitating the processing of non-uniformly sampled time domain data. As will be shown in the paper, the unique features of WB provide it with novel abilities to enhance the quality, accuracy, and fidelity of the spectral reconstruction process. WB also provides features which promote collaboration, education, parameterization, and non-uniform data sets along with processing integrated with the Rowland NMR Toolkit (RNMRTK) and NMRPipe software packages. WB is available free of charge in perpetuity, dual-licensed under the MIT and GPL open source licenses.

Iterative Development of an Application to Support Nuclear Magnetic Resonance Data Analysis of Proteins

The CONNecticut Joint University Research (CONNJUR) team is a group of biochemical and software engineering researchers at multiple institutions. The vision of the team is to develop a comprehensive application that integrates a variety of existing analysis tools with workflow and data management to support the process of protein structure determination using Nuclear Magnetic Resonance (NMR). The use of multiple disparate tools and lack of data management, currently the norm in NMR data processing, provides strong motivation for such an integrated environment. This manuscript briefly describes the domain of NMR as used for protein structure determination and explains the formation of the CONNJUR team and its operation in developing the CONNJUR application. The manuscript also describes the evolution of the CONNJUR application through four prototypes and describes the challenges faced while developing the CONNJUR application and how those challenges were met.

An overview of the BioExtract Server: a distributed, Web-based system for genomic analysis

Genome research is becoming increasingly dependent on access to multiple, distributed data sources, and bioinformatic tools. The importance of integration across distributed databases and Web services will continue to grow as the number of requisite resources expands. Use of bioinformatic workflows has seen considerable growth in recent years as scientific research becomes increasingly dependent on the analysis of large sets of data and the use of distributed resources. The BioExtract Server (http://bioextract.org) is a Web-based system designed to aid researchers in the analysis of distributed genomic data by providing a platform to facilitate the creation of bioinformatic workflows. Scientific workflows are created within the system by recording the analytic tasks preformed by researchers. These steps may include querying multiple data sources, saving query results as searchable data extracts, and executing local and Web-accessible analytic tools. The series of recorded tasks can be saved as a computational workflow simply by providing a name and description.